3 The Underlying Causes I have suggested that the immediate causes of the depression were the confluence of risky lend- ing with inadequate personal savings, so that when the risks materialized, causing bank insolvencies and a fall in demand for goods and services be- cause credit was difficult to obtain, people couldn’t reallocate savings to consumption, and this al- lowed the fall in demand to trigger a downward spiral in employment and output. Digging a little deeper, we find the housing bubble, the bursting of which produced the defaults that endangered the solvency of the banks; the very low interest rates that motivated the banks to increase their leverage; the complicated financial instruments that turned out to be riskier than people thought; and the with- ering of regulation of financial services, which re- moved checks on risky lending. (The bursting of the housing bubble also of course caused the resi- 75 A Failure of Capitalism dential-construction market to nose-dive, but it is the broader decline in demand for goods and ser- vices that poses the big threat to the economy.) These phenomena too need to be explained rather than just assumed, if there is to be hope of heading off future depressions, and of a prompt recovery from this one. Among the deep causes of a depression might be human errors—maybe errors of a kind to which people are predisposed by quirks of human cognitive psychology—or character flaws, such as “greed” (whatever that means). It is widely be- lieved, for example, that banks miscalculated the safety of the novel financial instruments, or were taken in by the credit-rating agencies’ giving triple- A ratings to what really were risky assets; and that the housing bubble was due to the inability of ordinary people to understand the risks involved in novel forms of mortgage loan, such as a mortgage loan that does not require the borrower to have any equity in his home and does not give him the secu- rity of a fixed interest rate. It has been suggested that bubbles reflect irrational optimism or perhaps an inability even of financial professionals to base predictions on anything more sophisticated than simple extrapolation, so that if house values or 76 The Underlying Causes stock values are rising, the financial markets as well as the consuming public expect them to keep rising indefinitely. One even hears it said that financiers are stupid. I am skeptical that readily avoidable mistakes, failures of rationality, or the intellectual deficiencies of financial managers whose IQs exceed my own were major factors in the economic collapse. Had the mistakes that brought down the banking indus- try been readily avoidable, they would have been avoided. There were plenty of warnings of a hous- ing bubble, beginning in 2003; warnings about ex- cessive leverage in financial firms; and even rather precise predictions of the debacle that has ensued, as in “When Bubbles Burst,” an eerily prescient paper by Thomas Helbling and Marco Terrones published in October 2003. Robert Shiller wrote a similar paper in April of the following year, as did another economist, Avinash Persaud, the same month. The Economist magazine spotted the hous- ing bubble in September 2002 and soon became obsessed with it and its possible broader implica- tions for the financial system and the U.S. econ- omy as a whole; in an article published on July 3, 2004, we read: “Housing optimists dismiss these fears by pointing out that doomsters such as The 77 A Failure of Capitalism Economist began wringing their hands about a property bubble a year ago, and yet prices have continued to climb. But this has made the housing market not safer, but more vulnerable. The first law of bubbles is that they inflate for a lot longer than anybody expects. The second law is that they eventually burst.” A particularly good piece, called “Will It Be Different This Time?,” ran in the Econ- omist in October of 2004. The Financial Times soon picked up the thread. The fragility of the banking system and the inadequacy of banking regula- tion were related topics discussed frequently in the financial press. The banks must have known as much as econo- mists and financial journalists did about their in- dustry. They had to know that there was a lot of risk in their capital structures, that the future doesn’t always repeat the past and therefore that models of default risk based on historical experience in the housing and credit markets might be unreli- able, that credit-rating agencies have a conflict of interest because they’re paid by the firms they rate, and that financial intermediation is inherently un- stable because to be profitable it usually requires borrowing short-term and lending long-term. So what were the bankers to do? In gauging the risk of 78 The Underlying Causes calamity, the key probabilities they had to consider were that the rise in housing prices was a bubble and that if it burst house prices would fall by at least 20 percent. If both events came to pass, insol- vency would loom. Suppose the best guess was that there was a 10 percent probability that the price rise was a bubble and the same probability that if it was a bubble house prices would fall by at least 20 per- cent. Then the probability that house prices would fall by at least 20 percent was only 1 percent (0.1 × 0.1), and so disaster would be unlikely to occur for many years, and so the risk of disaster would have seemed worth running. A 1 percent risk of bank- ruptcy is not like a 1 percent risk of a nuclear war. Bankruptcy is common enough, in fact is an indis- pensable institution of a capitalist society. Because risk and return are positively correlated, a firm that plays it too safe is, paradoxically, courting failure because investors will turn elsewhere. Businesses must assume a positive though small risk of bank- ruptcy. A cascade of bank bankruptcies can bring the economy to a halt, but no individual bank has an incentive to take measures to avoid such a con- sequence. We need to consider intrafirm conflicts of inter- est if we want to obtain a better understanding of 79 A Failure of Capitalism how the rational decisions of intelligent people can lead to disaster. Banks employ risk managers as well as traders (I include in the term “traders” loan officers and anyone else who makes lending or in- vestment decisions for the bank). But the two types of employee have inherently conflicting objec- tives—profit for the traders, safety for the risk man- agers—and these are reciprocals: more profit, less safety. Risk management, unlike trading, is gener- ally not treated as a profit center in a firm, because it is difficult to attribute profits to risk managers, just as it is difficult to attribute profits to the firm’s accountants and lawyers, who also are risk manag- ers in effect. Hence a financial firm will tend to give more weight to the views of successful traders than to those of risk managers. With the rapid expansion of the financial sector in the early 2000s, young, inexperienced traders and analysts achieved positions of responsibility in banks before they were fully prepared by train- ing and experience. And when an organization ex- pands rapidly there is bound to be some loss of control over subordinates, which exacerbates the problem of having a less experienced staff. This is not to blame the financial crisis on the young. Of- ten, perhaps typically, the decision to ramp up the 80 The Underlying Causes level of risk in a bank’s loan and investment port- folios came from on high, as when Robert Rubin persuaded the rest of the senior management of Citigroup to increase the company’s return by tak- ing more risk. In a well-managed company (no one thinks Citigroup well managed), push-back from below would be encouraged—from brilliant ana- lysts more knowledgeable than senior managers about the limitations of the impressive mathemati- cal tools of risk management. My point is only that problems of communication and control are en- demic to organizations, especially large ones, and played a role in the crisis. The emergence of the organizational problems that I have mentioned coincided with the creation of the new financial instruments—the mortgage- backed securities and credit-default swaps and other novel financial instruments that were even more complex. Organizations stressed by rapid expansion were further stressed by the analytical challenges posed by the new instruments, just as the financial institutions of the 1920s had been stressed by the new financial methods of that era, such as installment buying by consumers and bank lending to buyers of stock on margin. Organiza- tional problems (another one is the tendency I 81 A Failure of Capitalism mentioned to weight the advice of traders more heavily than that of risk managers) increase the likelihood of mistakes. But they are problems to avoid or solve; they do not signify irrationality. Emotion does play a role in the behavior of busi- nessmen and consumers, as of all human beings, but it is not necessarily or even typically irrational. It is a form of telescoped thinking, like intuition, and often it is superior to conscious analytic proce- dures. A driver who swerves in alarm to avoid hit- ting a pedestrian is not irrational because he failed to conduct a cost-benefit analysis before swerving. Many of the examples given to show that financial behavior is irrational are superficial—for example that many people buy stocks at the peak of the mar- ket and sell at the trough, when they should be doing the reverse.